Pipeline commands

Apothem ships an architecture-aware /plan pipeline, a thirteen-stage /research pipeline, plus an eleven-command audit sequence — giving every project a structured path from raw prose to production-grade, multi-axis-reviewed code, and from a raw idea to a disseminated, peer-reviewed paper.

Pipeline overview

flowchart LR
    accTitle: Plan pipeline command flow
    accDescr: Left-to-right flowchart of the apothem plan pipeline from the spec stage through the generate and review stages, then through a conditional architecture-bearing decision that either runs the design stage or skips directly to the execute stage. The orthogonal audit stage wraps the spec-through-review span as a closed remediation loop.
    A[/plan-spec] --> B[/plan-generate]
    B --> C[/plan-review]
    AUD[/plan-audit<br/>orthogonal: wraps spec → review] -.->|closed-loop remediation| A
    C --> Q{Architecture-bearing suite?}
    Q -->|yes| F[/plan-design<br/>emit _inputs/design.md]
    Q -->|no| D[/plan-execute]
    F --> D
    D --> E[/plan-status]
    D --> G[Audit review sequence]
    G --> H[Completion milestone]

The planning stages

The pipeline ships as eight first-class slash commands, one per stage — each independently invocable as /plan-<stage>.

/plan-audit is orthogonal to the linear flow, not a stage within it. It wraps /plan-spec → /plan-generate → /plan-review as a single closed loop, runs only the missing upstream stages, actively remediates findings, and leaves the suite ready for /plan-execute. It is distinct from the audit review sequence below, which audits an executed codebase rather than a plan suite.

/plan-design is not a universal ceremony. /plan-review classifies whether the suite changes architecture-bearing surfaces. If yes, /plan-design emits the design trace that /plan-execute and /architecture-review consume. If no, the reviewed suite moves directly into execution.

The research pipeline

The research pipeline ships as thirteen first-class slash commands, one per stage — each independently invocable as /research-<stage>, chained by a Sequence Gate and a Handoff Manifest, under ten rigor mandates (authoritative sources, reproducibility, falsifiability, citation integrity, preregistration, ethics / conflicts, statistical rigor, open-science / FAIR, reporting-guideline conformance, theoretical grounding / impact).

flowchart LR
    accTitle: Research pipeline command flow
    accDescr: Left-to-right flowchart of the apothem research pipeline from the ideate stage through spec, theory, sources, synthesis, proposal, study design, experiment, analysis, paper, and peer review to publish and disseminate, with a revision loop from review back to paper.
    RI[/research-ideate] --> RS[/research-spec]
    RS --> RT[/research-theory]
    RT --> RSO[/research-sources]
    RSO --> RSY[/research-synthesis]
    RSY --> RPR[/research-proposal]
    RPR --> RD[/research-design]
    RD --> RE[/research-experiment]
    RE --> RA[/research-analysis]
    RA --> RP[/research-paper]
    RP --> RR[/research-review]
    RR --> RPU[/research-publish]
    RPU --> RDI[/research-disseminate]
    RR -.->|revision loop| RP

The /research wrapped-workflow orchestrator drives these stages from raw question to published paper (it is not itself a stage), checking each stage's output independently before the next stage builds on it; --quick runs /research-sources + /research-synthesis only for a single-shot cited report.

The empirical stages (sources, design, experiment, analysis) and the writing stages (paper, review, publish, disseminate) additionally draw on operationalizing rigor surfaces — fair multi-method comparison, comparator provenance, an experiment-program scaffold, a compute-utilization plan, an opt-in autonomous experiment loop, a definition-of-done advancement gate, and blinding with staged disclosure — that add operational detail to the ten mandates without extending them.

Audit review sequence

The /audit wrapped-workflow orchestrator sweeps all eleven dimensions below as a single parallel pass — the audit-fortress analogue of /plan and /research — synthesizing one deduplicated, severity-triaged findings report (report-only; remediation routes to /elevate). Each dimension is also individually invocable.

The /fortress wrapped-workflow orchestrator closes the loop /audit reports: it detects through /audit, verifies each finding, remediates every survivor at its owning surface, re-audits in a bounded loop, and gates the result through /release-readiness — the production-hardening analogue of /plan and /research.